Skip to main content

Responses are generated using AI and may contain mistakes.

CollectorsCloud Collectors Administration Guide

Prerequisites to Configure the ChatGPT Enterprise Cloud Collector

Before you configure the ChatGPT Enterprise cloud collector, you must complete the necessary prerequisites. To begin, ensure that you have owner permissions for both the OpenAI Platform and ChatGPT Enterprise.

The prerequisite procedures in the subsections below will enable you to gain access to the ChatGPT Compliance API and to obtain information needed for configuring the cloud collector.

Specifically, you will:

  • Locate the Organization ID in both your OpenAI Platform and your ChatGPT Enterprise, and ensure that the IDs match. If these IDs do not match, contact OpenAI Support at [email protected].

  • Obtain the Workspace ID of your ChatGPT Enterprise data source.

  • Create an OpenAI Platform Admin key. You will need this key to ensure your ChatGPT Enterprise cloud collector can access the ChatGPT Compliance API.

Locate ChatGPT Enterprise Workspace Details

  1. Navigate to your ChatGPT Enterprise workspace.

  2. In the lower left corner, click the user icon and select the Workspace settings option. The General page opens.

  3. From the Workspace details section, copy both of the following IDs:

    • Organization ID – You will need this ID for comparison with the Organization ID from the OpenAI Platform to ensure they match.

    • Workspace ID – You will need to provide this ID when contacting OpenAI Platform in the next procedure and when configuring the cloud collector itself.

    chatgpt-general.png

Locate OpenAI Platform Workspace Details and Keys

  1. Navigate to OpenAI Platform. If prompted, choose Continue with Google.

  2. In the left navigation pane, navigate to Settings and select General.

    Note

    Ensure that you have logged in to your organization workspace and not your own personal workspace.

  3. Copy the Organization ID. Compare this ID with the Organization ID in your ChatGPT Enterprise to ensure they match. If these IDs do not match, contact OpenAI Support at [email protected].

    openai-general.png

  4. In the left navigation pane, select Admin keys.

  5. Create a new key as follows:

    • Click Create new admin key. The Create new admin key dialog box opens.

    • In the Name field, enter a name for the new admin key.

    • In the Permissions section select the All option.

    • Click the Create admin key button. The OpenAI Platform generates your new key. Copy and securely save information from this key, including the Key name, the Secret key, and the Created by. You will need this information for the next step and for configuring the cloud collector.

    openai-create-admin-key.png

  6. Send an email to OpenAI support, at [email protected], to request that Compliance API access be enabled for your organization's API key. In the email, request Read and Delete scope and provide all of the following account details:

    • ChatGPT Workspace ID – Obtained in Step 3 of Locate ChatGPT Enterprise Workspace Details.

    • Open AI Platform Organization ID – Obtained in Step 3 of this section.

    • API Key – Provide the last four characters of the Secret key you obtained in Step 5.

    • Key Name – Obtained in Step 5.

    • Created by – Obtained in Step 5.

  7. When you receive approval from OpenAI, you can Configure the ChatGPT Enterprise Cloud Collector.

Configure the ChatGPT Enterprise Cloud Collector

Set up a ChatGPT Enterprise cloud collector to continuously ingest data from your ChatGPT Enterprise. This cloud collector supports data collection from the Compliance API of the ChatGPT Enterprise.

  1. Before you configure the ChatGPT Enterprise Cloud Collector, ensure that you complete the prerequisites.

  2. Log in to the New-Scale Security Operations Platform with your registered credentials as an administrator.

  3. Navigate to Collectors > Cloud Collectors.

  4. Click New Collector.

  5. Click the tile for the ChatGPT Enterprise cloud collector. A configuration panel opens on the right and the remaining steps in this procedure can be used to customize it to create a collector that ingests ChatGPT Enterprise data.

    chatgpt-enterprise.png

  6. In the Configuration section of the panel, enter the following information:

    • Name – Enter a name for the new ChatGPT Enterprise cloud collector.

    • Authorization Token – Enter the Secret Key that you acquired when you created a new OpenAI admin key during the prerequisites.

    • Workspace ID – Enter your organization's ChatGPT Workspace ID. You acquired this ID during the prerequisites.

    • Data Sources – From the drop down menu, select the Conversations option to ingest conversations from the ChatGPT Compliance API.

  7. (Optional) SITE – Select an existing site or to create a new site with a unique ID, click manage your sites. Adding a site name helps you to ensure efficient management of environments with overlapping IP addresses.

    By entering a site name, you associate the logs with a specific independent site. A sitename metadata field is automatically added to all the events that are going to be ingested via this collector. For more information about Site Management, see Define a Unique Site Name.

  8. (Optional) TIMEZONE – Select a time zone applicable to you for accurate detections and event monitoring.

    By entering a time zone, you override the default log time zone. A timezone metadata field is automatically added to all events ingested through this collector.

    Timezone_sitename_site_management_1.png

  9. To confirm that the New-Scale Security Operations Platform communicates with the service, click Test Connection. 

  10. Click Install.

    AWS_S3_2.png

    A confirmation message informs you that the new Cloud Collector is created.

Note

Note the following potential issue of delayed data ingestion due to OpenAI Compliance API.

If you notice that data ingestion by the ChatGPT Enterprise cloud collector has been stopped, contact OpenAI support to confirm if their Compliance API is running correctly.

Recommendation – Contact OpenAI support to investigate and remediate the ingestion delay or interruption.